[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
xod:
> On Thu, 7 Aug 2003, John Cowan wrote:
>
> > seqram2 scripsit:
> >
> > > Instead of saying that Mr. Bird is "sometimes" carnivore and sometimes
> > > herbivore, sometimes male and sometimes female, it would seem to me
> > > that what makes for a "prototype" are exactly those characteristics
> > > which *are* constant across "almost all" instances of the critter. So
> > > Mr. Bird has no specific size (though its size is limited to between
> > > an inch or so and 10-odd feet), nor a gender, nor a diet, but does
> > > have feathers, a beak, two wings, is warm-blooded, etc.
> >
> > You are talking about lo'e cipni (in the non-Jorge sense of lo'e), which
is
> > indeed "thin" in the sense of having only those properties shared by
all,
> > or at least all typical, birds. Mr. Bird is a different thing: he is
the
> > One Bird of whom all individual birds are avatars. Whether English (or
other
> > SAE languages) take Mr. X or individual Xes to be more important
> > depends on X: for birds, it's the individual bird most of
> > the time, but for the _New York Times_ it's Mr. Times that counts, not
> > a particular copy/individual/avatar.
>
> But when we think of all those newspapers as being Mr. Times, it's ONLY
> because we are deliberately ignoring the individual differences. In that
> sense, we only invoke Mr. Times when we seek the Prototype (or Prototype
> Instance) that is related to CLL-lo'e.
We're not ignoring the individual differences. Rather, we don't treat those
differences as criteria for distinguishing one individual from another.
Rather,
they're seen as different states of the same individual.
And we can certainly say "John is reading the NY Times" without seeing that
as in any way characteristic of the NY Times. (Though as it happens, in
fact,
"read by John Cowan" is for me one of the most salient characteristics of
the NY Times.)
> The difference between Prototype and Prototype Instance is that statements
> about the former are definitional, while statements about the latter are
> not. However, definitional statements can (and should?) be made using
> "ro".
It would be useful to have something that is a blend of ro and so'e, for
making default generalizations. Something weaker than ro, but a little
stronger than the overly-statistical so'e.
> And we only need the Prototype Instance when we seek interchangeable
> members, which is probably the only use for lo.
I'm not sure what you mean by "lo". In only {lo broda}? In {su'o lo
broda}? In {su'o broda}? In {re lo mu broda}?
--And.